on

Can the Apple M2 (16 GB) run Phi-3.5 Vision Instruct?

36Basic quality (D) · 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.

Yes, it runs well

Runs well. At Q4_K_M it needs about 7.2 GB of 10.7 GB usable, with an estimated speed of 25.5 tok/s.

Best quant
Q4_K_M
Generation speed
25.5 tok/s
Prompt processing
538 tok/s
First token (1K prompt)
1.9 s
Largest safe context
16K

Best app to run it on the Apple M2 (16 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX · use the MLX build

  • Runs the MLX build, typically 20–35% faster than GGUF for a model with 4.2B active parameters on Apple silicon.
  • Point-and-click: search, download and chat in one window.

Open Phi-3.5 Vision Instruct in LM Studio

  • MLX LM Runs the MLX build, typically 20–35% faster than GGUF for a model with 4.2B active parameters on Apple silicon.
  • Jan Runs the MLX build, typically 20–35% faster than GGUF for a model with 4.2B active parameters on Apple silicon.

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Phi-3.5 Vision Instruct

Original weights · Hugging Face

Full-precision files from Microsoft, for fine-tuning, vLLM, Transformers.

huggingface.co/microsoft/Phi-3.5-vision-instruct (opens in a new tab)

No sign-in needed.

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 36Basic quality (D) · 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.
How capable the model itself is at Q4_K_M, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs well
How well it runs on the Apple M2 (16 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.16.2 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.12.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.15.0 / 15
Match score68.2 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the Apple M2 (16 GB)

Quant Memory needed Available Speed Runs Quality Download
Q2_K 6.3 GB 10.7 GB 41.8 tok/s Runs great 26Basic quality (D) · 26 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q3_K_M 6.7 GB 10.7 GB 32.9 tok/s Runs great 31Basic quality (D) · 31 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q4_K_M 7.2 GB 10.7 GB 25.5 tok/s Runs well 36Basic quality (D) · 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q5_K_M 7.6 GB 10.7 GB 21.0 tok/s Runs well 36Basic quality (D) · 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q6_K 8.1 GB 10.7 GB 17.7 tok/s Runs well 37Basic quality (D) · 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
Q8_0 9.1 GB 10.7 GB 13.0 tok/s Runs well 38Basic quality (D) · 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. All files (opens in a new tab)
F16 13.0 GB 10.7 GB 5.1 tok/s Slowfar over GPU memory limit 38Basic quality (D) · 38 of 100: how capable the model is at F16, whatever it runs on. Match score 41/100 (quality, fit, speed and memory use) sets the ranking. Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

Quick answers

How much memory does Phi-3.5 Vision Instruct need on the M2 16GB?

At Q4_K_M it needs about 7.2 GB. The M2 16GB has 10.7 GB usable.

How fast is Phi-3.5 Vision Instruct on the M2 16GB?

About 25.5 tok/s at Q4_K_M, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Phi-3.5 Vision Instruct should I use on the M2 16GB?

Q4_K_M. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.

Was this page helpful?